TY - GEN
T1 - An Evaluation of Software Frequency Sketches
AU - Friedman, Roy
N1 - Publisher Copyright: © 2024 Owner/Author.
PY - 2024/6/24
Y1 - 2024/6/24
N2 - Sketches enable maintaining statistics about very large streams of data in a memory frugal manner. This work presents a detailed evaluation of software implementations of several popular sketching solutions, as well as recently proposed optimizations. We compare these solutions in terms of computational speed, memory consumption, and several approximation error metrics. Overall, we find that a simple hashing based solution employed with the Nitro sampling technique [23] gives the best trade-off between memory, error and speed. Our findings also include some novel insights about how to best combine sampling with Counting Cuckoo filters depending on the application.
AB - Sketches enable maintaining statistics about very large streams of data in a memory frugal manner. This work presents a detailed evaluation of software implementations of several popular sketching solutions, as well as recently proposed optimizations. We compare these solutions in terms of computational speed, memory consumption, and several approximation error metrics. Overall, we find that a simple hashing based solution employed with the Nitro sampling technique [23] gives the best trade-off between memory, error and speed. Our findings also include some novel insights about how to best combine sampling with Counting Cuckoo filters depending on the application.
UR - http://www.scopus.com/inward/record.url?scp=85200654395&partnerID=8YFLogxK
U2 - 10.1145/3629104.3666028
DO - 10.1145/3629104.3666028
M3 - منشور من مؤتمر
T3 - DEBS 2024 - Proceedings of the 18th ACM International Conference on Distributed and Event-Based Systems
SP - 18
EP - 29
BT - DEBS 2024 - Proceedings of the 18th ACM International Conference on Distributed and Event-Based Systems
T2 - 18th ACM International Conference on Distributed and Event-Based Systems, DEBS 2024
Y2 - 25 June 2024 through 28 June 2024
ER -